© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 14664
    Prev
    Next
    INDEX
    Explanations

    keywords relating to manuals and instructions.

    oai_token-act-pair · gemini-2.0-flash

    keywords related to programming and coding instructions

    np_token-act-pair-logits · gpt-4o-mini

    instructions and directions

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Instruction
    -1.29
     instruction
    -1.21
    Instruction
    -1.14
    instruction
    -1.09
     Instructions
    -1.09
     instructions
    -1.05
     INSTRUCTION
    -1.04
     instructing
    -0.98
    Instructions
    -0.96
    instructions
    -0.96
    POSITIVE LOGITS
    WriteLiteral
    0.52
    ISupport
    0.49
    jetbrains
    0.44
    simple
    0.44
    nullable
    0.42
     uLocal
    0.41
    ulism
    0.41
     dating
    0.40
     simple
    0.40
    basic
    0.40
    Activations Density 1.030%

    No Known Activations